In a cellular telecommunication system, a plurality of user terminals within a cell communicate with a primary station. With the subsequent generations of cellular systems, the achievable data rate has been keeping increasing. In advanced systems such as UMTS and LTE, multi-antenna transmission/reception techniques variously described as, MIMO, precoding or beamforming are supported for transmissions from a single cell to a mobile terminal. Thanks to the spatial selectivity of the beamforming mode, such transmission modes have enabled an important increase of the achievable data rate and of the range of communication, while maintaining the average interference level.
In order to achieve the beamforming, a typical transmitting station having an antenna array applies a set of complex coefficients (forming a precoding matrix or precoding vector) to a signal transmitted from its respective antennas, so that the transmission stream is spatially directed towards a receiving station. However, reception of such a beamformed transmission may require the communication of this set of complex coefficients between the transmitting station and the receiving station. In implementations of such systems, precoding codebooks are defined. These precoding codebooks may be viewed as a way of describing precoding matrix (or precoding vector) of the channel coefficients or precoding weights in a compact way, thereby reducing the amount of required signalling for indicating the precoding. These codebooks also enable the user terminal (defined in LTE as a User Equipment or UE) to report to the network a preferred precoder for downlink transmission, in the form of an index to codebook entry. In this case, the preferred precoder is a set of complex coefficients to be applied to transmit antennas of the base station (defined in LTE as an eNodeB). Similarly, precoding codebooks may also be used by the base station to signal the precoding used for a transmission to the user terminal. This enables the user terminal to derive an appropriate phase/amplitude reference from common pilot symbols for demodulation of each downlink transmission.
In LTE, this signalled codebook index is referred to as PMI (Precoding Matrix Indicator). Typically, the same codebook may be used on one hand to signal to a user terminal the precoding vector or matrix which is actually applied in the downlink by a base station and on the other hand to feed back the preferred precoding matrix by the user terminal to enable a phase/amplitude reference to be derived. Alternatively, the reference(s) may be provided by precoded reference symbols (i.e. dedicated reference symbols).
Recently, it has been proposed to use cooperative beamforming, i.e. beamforming using antennas from multiple cells or multiple base station sites (under the description of CoMP or Co-operative Multi-Point transmission). Such systems are introduced on FIGS. 1 to 3. In such a system, a user terminal 110 within a serving cell 101a communicates in normal (i.e. single cell) mode with a primary station 100a. In normal beamforming mode, the primary station 100a applies a set of precoding weights to the signal to be transmitted from its transmit antennas 104a to create a spatial stream 105a directed towards the user terminal 110.
In a cooperative beamforming mode, a second primary station 100b in a neighboring cell 101b uses some of its antennas 104b to transmit in a cooperative way the same signal 105b as the one 105a transmitted by the primary station 100a to the user terminal 110. The spatial stream now comprises two components 105a and 105b. As explained above, the user terminal 110 needs to feed back a channel state estimate based on measurements on received reference symbols. This estimate in this example is an indication of a preferred precoding matrix (or vector if there is only one transmission stream) in the form of a codebook index.
As illustrated on FIG. 1, it may be possible to report a PMI for each co-operating cell, i.e. the user terminal signals transmits an indication of a first preferred precoding matrix 111a for the serving cell 101a to the primary station 100a and an indication of a second preferred precoding matrix 111b for the cell 101b to the primary station 100b. Thus, the base stations 100a and 100b may use different precoding in order to have a fine adjustment of the transmission beam. This means that the user terminal needs to feed back as many PMIs as there are cooperating cells. This may represent a great amount of signalling and overhead.
In order to reduce this signalling, as illustrated on FIG. 2, it could be possible that the user terminal 110 transmits only one PMI to the cooperating cells base stations 100a and 100b. On FIG. 2, the user terminal 110 makes an estimate of the received transmission channels by means of measurements on reference symbols, and establishes one PMI 112 which is transmitted to both base stations 100a and 100b. This means that the base station 100a and the base station 100b apply the same precoding. Thus, the drawback of this is a lack of flexibility. Moreover, in case of more than two cooperating cells, it may be difficult to obtain an efficient beamforming.
Another approach illustrated on FIG. 3, may be described as SFN with antenna selection. Here SFN is “single frequency network”, which implies that the same signal is transmitted from more than one cell. The user terminal 110 reports a value of precoding matrix 113a which is to be applied by all the co-operating cells 101a and 101b. Additionally the user terminal 110 signals in a further signalling message 113b whether particular antennas should be switched off. This means that the remaining antennas are “selected”. In a typical implementation of this technique for the case of 4 transmit antennas per cell, up to one antenna may be switched off per cell. The user terminal then needs to search the possible combinations of PMI and antenna selection for the one which will give the highest data rate.
This permits some more flexibility in the adjustment of the precoding. However, the antenna selection feedback needs some data. For instance, in the above example, the antenna selection for one cell needs at least 3 bits (four different antenna values and the case where no antenna is switched off). This means that this causes overhead and still a significant amount of signalling. Moreover, this may lead for some base stations to a potential power imbalance between the different antennas. This may thus reduce the total available power to that provided by the remaining on antennas, and this would affect the achievable transmit data rate.